Short report

Histopathological evaluation of scleritis

J Clin Pathol: first published as 10.1136/jclinpath-2018-205360 on 5 February 2019. Downloaded from http://jcp.bmj.com/ on 8 February 2019 by guest. Protected by copyright.
Mark Hankins,1 Curtis Edward Margo‍ ‍ 1,2

1
Department of Ophthalmology, Abstract the classification and histopathological features of
USF Health Morsani College of The sclera is an uncommon site of primary inflammation. scleritis, and the role that biopsy plays in the evalu-
Medicine, Tampa, Florida, USA
2
Department of Biopsy is infrequently employed in the evaluation of ation of scleritis.
Dermatopathology, USF Health scleritis, but familiarity with its differential diagnosis is
Morsani College of Medicine, instrumental in ensuring efficient histological evaluation. Clinical context
Tampa, Florida, USA This review provides a clinical overview of scleritis The epidemiology of scleritis is incomplete and
and describes the context in which scleral biopsy relies mainly on data from tertiary-care medical
Correspondence to might arise. Most cases are associated with systemic
Dr Curtis Edward Margo,
centres.1–5 Onset is usually in late middle age (45
Dermatopathology, USF Health autoimmune disease, but a sizeable proportion occur as to 60 years), rarely occurs during childhood and
Morsani College of Medicine, an isolated disorder. Conditions mimicking autoimmune affects women more than men.1–5
Tampa 33612, Florida, USA ; ​ scleritis include infection and neoplasm. Histological The clinical classification of scleritis is based on
cmargo@​health.​usf.​edu patterns of inflammation in eyes removed surgically region of the eye affected (anterior or posterior
Received 29 June 2018
or at autopsy have been placed into three groups: (1) (behind the insertion of the rectus muscles), or
Revised 6 January 2019 autoimmune scleritis characterised by varying mixtures both), whether inflammation is nodular or diffuse,
Accepted 14 January 2019 of palisading granulomas, necrosis and vasculitis; (2) and the presence or absence of necrosis. This
infectious scleritis, characterised by acute inflammation scheme has not substantially changed since it was
and necrosis; and (3) idiopathic scleritis, characterised first proposed in the 1960s.3 6
by chronic non-specific inflammation with follicles and In terms of aetiological diagnoses, roughly half of
varying amounts of fibrosis. This traditional system of patients with scleritis have an underlying systemic
classification may be oversimplified. Aetiological or disease, most of which are autoimmune.1–3 7 In
categorical classification is not always possible on small clinical series, the most common systemic disor-
biopsies given the histopathological overlap of infectious ders are rheumatoid arthritis, relapsing polychon-
and non-infectious scleritis. dritis, granulomatous polyangiitis and systemic
lupus erythematosus.1–3 7 Nearly all autoimmune or
collagen vascular diseases have been causally impli-
cated with scleritis at some time. Although the sclera
Introduction is an uncommon site of primary infection, infection
Scleritis describes inflammation of the outer collag- makes up a substantial proportion of cases found
enous tunic of the eye. Although scleral biopsy in systematic reviews of scleritis.3 When bacteria or
is not often called on in the evaluation of scleral fungi infect the sclera, there is usually a history of
inflammation, familiarity with the disorders causing accidental trauma or surgery. Unlike exogenously
scleritis should expedite efficient histological assess- acquired infections, Herpes zoster scleritis is a mani-
ment of tissue when biopsy is needed. The sclera festation of latent activation.3 Occult infections are
is the white tissue that transitions anteriorly at the uncommon, but when they occur, the organism can
limbus to clear cornea. It is sandwiched between be discovered on biopsy.8 Surgery can also induce
vascular episclera tissues and the uveal tract. The a non-infectious, necrotising scleritis, a diagnosis
sclera has no intrinsic vascular plexus of its own reached by exclusion.9 Scleritis can also be a side
and is an uncommon primary site of inflammation. effect of radiation therapy, including brachytherapy
In the absence of accidental or surgical trauma, for uveal melanoma, and has been attributed to the
the search for an aetiological diagnosis for scleritis use of bisphosphonates, both oral and parenteral.
involves identifying a systemic disease from a rela- Masquerade conditions that mimic scleritis are
tively limited number of tenable choices. Inflam- usually attributed to neoplasms, such as conjunc-
mation of the sclera can spill over to contiguous tival squamous cell carcinoma, lymphoma or uveal
tissues, in which situation terms like scleral keratitis melanoma.10 11
or scleral uveitis are applicable. The ocular compli-
cations of scleritis are potentially serious, including Histopathology
permanent vision loss and chronic unremitting pain. There are relatively few studies that have system-
© Author(s) (or their Scleral biopsy is not commonly performed for atically studied the pathology of scleritis. Larger
employer(s)) 2019. No diagnostic purposes due to the risks of worsening studies have relied on eyes removed surgically or
commercial re-use. See rights
and permissions. Published the inflammatory process or structurally weak- at autopsy, which could bias sampling to reflect
by BMJ. ening an already compromised tissue. Despite more advanced or severe forms of disease. The two
these concerns, biopsy is justified when conditions largest studies dealing with the histopathology of
To cite: Hankins M,
like infection or neoplasm cannot be otherwise scleritis were based on 87 eyes removed surgically
Margo CE. J Clin Pathol
Epub ahead of print: excluded. Insight into the morphological alter- or at autopsy and nine scleral biopsies.12 13 Investi-
[please include Day Month ations of scleritis comes mostly from eyes removed gators found that inflammation conformed to one
Year]. doi:10.1136/ surgically because they were blind and painful, or of three patterns, two of which displayed promi-
jclinpath-2018-205360 from eyes obtained at autopsy. This paper reviews nent necrosis. One group with scleral necrosis was
Hankins M, Margo CE. J Clin Pathol 2019;0:1–5. doi:10.1136/jclinpath-2018-205360    1
 Short report

J Clin Pathol: first published as 10.1136/jclinpath-2018-205360 on 5 February 2019. Downloaded from http://jcp.bmj.com/ on 8 February 2019 by guest. Protected by copyright.
Figure 1  Eye removed with Pseudomonas scleritis diagnosed with Figure 3  Granulomatous polyangiitis (diagnosed clinically with
positive cultures. The eye shows episcleral and outer scleral acute ANCA-positive serology) in a blind eye. Half the sclera and choroid
inflammation and necrosis. This region of sclera overlies the ciliary body (*) are effaced by inflammation. The lens and optic nerve are labelled
(CB) (H&E; original magnification ×120). Inset reveals necrotic sclera for orientation (H&E, original magnification ×5). Inset shows acute
with nuclear debris (H&E, original magnification X200). inflammation, multinucleated giant cells and necrosis (H&E, original
magnification ×200). ANCA, antineutrophil cytoplasmic antibody.

attributed to infection, corroborated with positive cultures or

identification of micro-organisms with histochemical stains immune-mediated disorders. The third group consisted of
(figures 1 and 2). Necrosis was also a prominent feature of auto- non-specific chronic inflammation, consisting of lymphocytes,
immune scleritis, in which devitalised collagen was surrounded plasma cells and histiocytes without substantial necrosis. Gran-
by varying proportions of neutrophils and palisading histiocytes ulation tissue, occasional lymphoid follicles and fibrosis were
(figures 3 and 4). Cultures and stains for micro-organisms were prominent in some cases (figure 5). Although occasional giant
uniformly negative. Giant cells and vasculitis were present in
some cases, but were often most notable in extrascleral loca-
tions. Essentially, all patients in this group had clinical or labo-
ratory evidence of immune-mediated disease, like rheumatoid
arthritis or granulomatous polyangiitis. Histopathological find-
ings alone were not sufficient to distinguish between different

Figure 4  Blind eye removed with rheumatoid scleritis diagnosed

clinically in a patient with advanced extra-articular disease. Upper panel
shows extensive patch of necrosis involving organised choroid and inner
sclera. Central sclera (S) is acellular. Outer sclera adjacent to tendon
Figure 2  Fungal scleritis. Scleral biopsy from a patient treated of medial rectus muscle (*) is inflamed (H&E, original magnification
unsuccessfully for months for presumed immune-mediated anterior ×70). Lower left shows sector of necrosis of inner sclera and choroid
scleritis. The sclera is filled with a myriad fungal hyphae (periodic acid– (H&E, original magnification ×120). Lower right shows an ill-defined
Schiff, original magnification X250). Inset shows clinical appearance palisading granuloma with multinucleated giant cells (H&E, original
prior to biopsy. magnification X200).
2 Hankins M, Margo CE. J Clin Pathol 2019;0:1–5. doi:10.1136/jclinpath-2018-205360
 Short report

J Clin Pathol: first published as 10.1136/jclinpath-2018-205360 on 5 February 2019. Downloaded from http://jcp.bmj.com/ on 8 February 2019 by guest. Protected by copyright.
Figure 6  Mycobacterial scleritis. Upper right, anterior scleritis
following vitrectomy that was initially considered immune mediated.
Figure 5  Eye removed for suspected choroidal melanoma shows thin Upper left shows biopsy of episclera with acute and chronic
choroid (arrowheads) and markedly thickened sclera. Overlying retina inflammation (H&E, original magnification ×200). Lower right is a
is not visible in this section. Collections of lymphocytes are scattered biopsy from sclera showing necrosis and nuclear debris (H&E, original
throughout the sclera, without appreciable necrosis (H&E, original magnification ×200). An acid-fast stain from area shown in upper right
magnification ×45). Inset shows collections of lymphocytes in sclera reveals rod-shaped organisms (probably not sufficiently decolourised);
and choroid. Once called brawny scleritis, this entity would now be cultures grew Mycobacterium chelonae (original magnification ×400).
evaluated for IgG4-related disease (H&E, original magnification X125).

melts.8 19 21 When blind or painful eyes are removed surgically

cells were noted, granulomatous inflammation was not a major
after brachytherapy for uveal melanoma, they not uncommonly
feature. These patients had no evidence of systemic disease and
show acellular scleral collagen surrounded by non-specific
corresponded clinically to idiopathic scleritis.
chronic inflammation (figure 7). Depending on the time from
These two studies have been criticised for over simplifying
radiation exposure, the sclera may or may not be attenuated. To
classification.14 Contradictions in the correlative stratagem was
our knowledge, bisphosphonate-associated scleritis has not been
noted early on. Herpes zoster scleritis, for instance, could display
the subject of pathological study.
a zonal pattern of granulomatous inflammation characteristic
of immune-mediated injury.12 The system also did not accom-
modate conditions like sarcoidosis, characterised by non-necro- Immunopathology
tising granulomatous inflammation.15 Other investigators found Few studies have examined the composition of inflammatory
considerable overlap in the patterns of immune-mediated and cells in scleritis. One study found a greater average proportion
idiopathic inflammation, making the classification system of CD20+ lymphocytes (43%) among three eyes with auto-
impractical.16–18 Further, infectious scleritis may display chronic immune-mediated scleritis than in three eyes with idiopathic
inflammation (non-granulomatous or granulomatous) with rela- scleritis (17%) (p=0.06).25 In this small study, there were no
tively little necrosis (figure 6).19 One factor that may explain differences in CD4, CD8 and CD68 immunoreactive cells. In
these differing results is sampling. Surgical biopsies are typi- another study based on 25 biopsies of necrotising scleritis and
cally small and superficial, consisting of episclera or overlying five biopsies with recurrent non-necrotising scleritis, vascular
conjunctiva. For this reason, optimal use of tissue may include immune deposits were found in 93% of scleral specimens and
the preparation of unstained slides in the event that additional in 79% of overlying conjunctival biopsies.26 Conjunctival tissue
histochemical stains or immunohistochemistry are needed on a contained significantly more T cells, macrophages and B cells
small biopsy. compared with 13 normal controls obtained during routine
More recently, IgG4-related disease has emerged as a cause cataract surgery, while the sclera displayed increases in T cells
of scleritis.20 When clinical and histopathological features of and macrophages compared with four normal scleras obtained
these cases are scrutinised, they appear to represent a subset of from eye bank eyes.26 HLA-DR expression was increased in
previously reported idiopathic scleritis, including a fibrosing both conjunctiva and sclera compared with controls.26 To date,
variant referred to as brawny scleritis or sclerosing inflammatory immune complexes have been found in cases of scleritis asso-
pseudotumour in the past.20 ciated with vasculitis.27 Expression of tumour necrosis factor
The histopathological features of surgically induced scleral (TNF)-α and interleukin-2 have been documented in non-gran-
necrosis and radiation-associated scleritis are not well docu- ulomatous scleritis, while TNF-α alone has been reported in
mented.3 9 21–24 Surgical and radiation-related scleritis usually necrotising scleritis.25 TNF-α has also been found in tears of
come to clinical attention because of scleral thinning, often patients with necrotising scleritis. The results of immunopa-
described as ‘melts’, which deter biopsy.23 The literature on thology studies and alterations in the inflammatory cascade have
surgical induced scleritis is heavily weighed to reporting occult shed some light on pathogenesis, but lack practical diagnostic
infections rather than the histopathology of idiopathic scleral application at this time.
Hankins M, Margo CE. J Clin Pathol 2019;0:1–5. doi:10.1136/jclinpath-2018-205360 3
 Short report

J Clin Pathol: first published as 10.1136/jclinpath-2018-205360 on 5 February 2019. Downloaded from http://jcp.bmj.com/ on 8 February 2019 by guest. Protected by copyright.
Figure 8  Squamous cell carcinoma masquerade syndrome. A clinically
inflamed scleral nodule progressively worsened despite treatment with
corticosteroids. Biopsy showed a squamous cell carcinoma involving
superficial sclera. The tumour likely arose from conjunctiva. The
peripheral cornea (PC) is labelled for orientation. The ciliary body and
iris are necrotic (periodic acid–Schiff, original magnification ×45). Inset
Figure 7  Dome-shaped melanoma of peripheral choroid and ciliary
shows infiltrating squamous cell carcinoma. Some cells contain clear
body 14 months after brachytherapy. The blind, painful eye show
spaces in the cytoplasm that failed to stain with periodic acid–Schiff
necrosis of the posterior half of the melanoma and necrosis of overlying
and colloidal iron (periodic acid–Schiff, original magnification ×200).
retina and inner sclera (H&E, original magnification ×5). Inset shows
acellular inner sclera (S) surrounded by necrotic tumour above and
chronically inflamed outer sclera below (H&E, original magnification
×45). and idiopathic scleritis, these conditions should be considered
in the histopathological evaluation of any scleral biopsy. Special
stains for micro-organisms and immunohistochemical studies
Conclusions for potential pathogens should be anticipated. Diagnostic biop-
This brief review summarises the relatively limited literature on sies are typically small as surgical manipulation of the sclera can
the histopathology of scleritis. Most scleritis can be categorised structurally weaken the eye. The judicious use of a small biopsy
based on clinical and laboratory findings as immune-mediated might include preparing unstained slides during initial sectioning
(systemic disease associated), idiopathic, surgically induced, in the event ancillary studies are needed. Recognition of small
drug-related or infectious. Scleral biopsy is usually needed to aid numbers of neoplastic cells can be enhanced through the use
in diagnosing cases with worsening or atypical clinical courses.
of select immunohistochemical studies (eg, cytokeratin, S100,
Since occult infection (eg, fungus, atypical mycobacteria, etc)
etc). Small or superficial biopsies may not always be sufficient to
or unrecognised neoplasms (eg, squamous cell carcinoma,
distinguish autoimmune from idiopathic scleritis, or infectious
lymphoma, etc) (figure 8) can clinically mimic autoimmune
from non-infection scleritis.
The morphological study of scleritis has been hampered by the
rarity of surgical biopsy and the reliance on anecdotal experience
Take home messages 
to guide practice patterns. Perhaps the greatest limitation in eval-
uating this material is to weigh the findings from larger clinical
►► Primary inflammation of the sclera is uncommon. Most
series from tertiary care institutions with the literature of case
cases are associated with autoimmune disease. An etiologic
reports, which often provides conflicting or differing perspec-
diagnosis can be obtained in the majority of cases based on
tives of the subject.
clinical and laboratory evaluation without biopsy.
Future research into scleritis holds considerable opportunity
►► There is considerable overlap in the histopathologic features
given the large proportion of cases due to unknown cause or have
of autoimmune scleritis and infectious scleritis. A definitive
unresolved pathogenesis. One particular promising avenue of
histopathologic diagnosis may not always be possible in the
investigation is the use of untargeted metagenonomic next-gen-
absence of positive identification of microorganisms in tissue
eration gene sequencing.28 This emerging technology is capable
sections.
of identifying the genetic signatures of infectious organisms, both
►► Occult or unsuspected neoplasms such as uveal melanoma,
common and rare, from extremely small samples (eg, swab or
squamous cell carcinoma, and lymphoma can mimic scleritis
minute biopsy). Another potentially valuable means of studying
clinically. Scleral biopsy is valuable in establishing these
diagnoses. immune-mediated scleritis is through gene expression profiling
►► Scleral biopsies tend to be small and superficial as the scleral
in which deviations from normal molecular patterns of gene
tunic is normally thin and critical for the structural integrity of expression may shed insight into autoimmune mechanisms.29
the eye. When receiving a small biopsy, preparing unstained
Handling editor  Tahir S Pillay.
sections may be helpful in the event that supplemental stains
are needed. Contributors  Both authors have contributed to the literature review and writing of
this paper. Both have read the final draft and approved its submission.

4 Hankins M, Margo CE. J Clin Pathol 2019;0:1–5. doi:10.1136/jclinpath-2018-205360

 Short report
Funding  The authors have not declared a specific grant for this research from any 12 Rao NA, Marak GE, Hidayat AA. Necrotizing scleritis. A clinico-pathologic study of 41

J Clin Pathol: first published as 10.1136/jclinpath-2018-205360 on 5 February 2019. Downloaded from http://jcp.bmj.com/ on 8 February 2019 by guest. Protected by copyright.
funding agency in the public, commercial or not-for-profit sectors. cases. Ophthamology 1985;92:1542–8.
13 Riono WP, Hidayat AA, Rao NA. Scleritis: a clinicopathologic study 55 cases.
Competing interests  None declared.
Ophthalmology 1999;106:1328–33.
Patient consent for publication  Not required. 14 Mondino BJ,Discussion. Ophthalmology 1985;92:1548–9.
Provenance and peer review  Not commissioned; externally peer reviewed. 15 Hessen M, Eberhart C, Butler N, et al. Sarcoidosis presenting as scleral nodule. Ocular
Immun Inflam 2014;22:311–3.
16 Sevel D. Necrogranulomatous scleritis. Clinical and histologic features. Am J
Ophthalmol 1967;64:1125–34.
References 17 Hatsuda T-A, Tanaka J. Bilateral necrotizing scleritis. Am J Ophthalmol
1 Sainz de la Maza M, Molina N, Gonzalez-Gonzalez LA, et al. Clinical characteristics 1978;86:710–4.
of a large cohort of patients with scleritis and episcleritis. Ophthalmology 18 Young RD, Watson PG. Microscopical studies of necrotising scleritis. I. Cellular aspects.
2012;119:43–50. Br J Ophthalmol 1984;68:770–80.
2 McCluskey PJ, Watson PG, Lightman S, et al. Posterior scleritis: clinical features, 19 Margo CE, Pavan PR. Mycobacterium chelonae conjunctivitis and scleritis. Arch
systemic associations, and outcome in a large series of patients. Ophthalmology Ophthalmol 2000;118:1125–8.
1999;106:2380–6. 20 Karim F, de Hoog J, Paridaens D, et al. IgG4-related disease as an emerging cause of
3 Okhravi N, Odufuwa B, McCluskey P, et al. Scleritis.. Surv Ophthalmol scleritis. Acta Ophthalmol 2017;95:e795–6.
2005;50:351–63. 21 Doshi RR, Harocopos GJ, Schwab IR, et al. The spectrum of postoperative scleral
4 Homayounfar G, Nardone N, Borkar DS, et al. Incidence of scleritis and necrosis. Surv Ophthalmol 2013;58:620–33.
episcleritis: results from the Pacific ocular inflammation study. Am J Ophthalmol 22 Yamazoe K, Shimazaki-Den S, Otaka I, et al. Surgically induced necrotizing scleritis
2013;156:752–8. after primary pterygium surgery with conjunctival autograft. Clin Ophthamol
5 Lavric A, Gonzalez-Lopez JJ, Majumder PD, et al. Posterior scleritis: analysis of 2011;5:1609–11.
epidemiology, clinical factors, and risk of recurrence in a cohort of 114 patients. Ocul 23 Kaliki S, Shields CL, Rojanaporn D, et al. Scleral necrosis after plaque radiotherapy of
Immunol Inflamm 2016;24:6–15. uveal melanoma: a case-control study. Ophthalmology 2013;120:1004–11.
6 Watson P, Romano A. The impact of new methods of investigation and treatment on 24 Chaudhry IA, Liu M, Shamsi FA, et al. Corneoscleral necrosis after episcleral Au-198
the understanding of the pathology of scleral inflammation. Eye 2014;28:915–30. brachytherapy of uveal melanoma. Retina 2009;29:73–9.
7 Akpek EK, Thorne JE, Qazi FA, et al. Evaluation of patients with scleritis for systemic 25 Usui Y, Parikh J, Goto H, et al. Immunopathology of necrotising scleritis. Br J
disease. Ophthalmology 2004;111:501–6. Ophthalmol 2008;92:417–9.
8 Jeang LJ, Davis A, Madow B, et al. Occult fungal scleritis. Ocul Oncol Pathol 26 Fong LP, de la Maza MS, Rice BA, et al. Immunopathology of scleritis. Ophthalmology
2017;3:41–4. 1991;98:472–9.
9 Sainz de la Maza M, Foster CS. Necrotizing scleritis after ocular surgery. A 27 Wakefield D, Di Girolamo N, Thurau S, et al. Scleritis: immunopathogenesis and
clinicopathologic study. Ophthalmology 1991;98:1720–6. molecular basis for therapy. Prog Retinal and Eye Res 2013;35:44–62.
10 Burton BJ, Cunningham ET, Cree IA, et al. Eye involvement mimicking scleritis in a 28 Gu W, Miller S, Chiu CY. Clinical metagenomic next-generation sequencing for
patient with chronic lymphocytic leukaemia. Br J Ophthalmol 2005;89:775–6. pathogen detection. Annu Rev Pathol 2019;14:317–36.
11 Sharma M, Sundar D, Vanathi M, et al. Invasive ocular surface squamous neoplasia 29 Li Z, Liu B, Maminishkis A, et al. Gene expression profiling in autoimmune
masquerading as nodular scleritis. Ophthal Plast Reconstr Surg 2017;33:e45–7. noninfectious uveitis disease. J Immunol 2008;181:5147–57.

Hankins M, Margo CE. J Clin Pathol 2019;0:1–5. doi:10.1136/jclinpath-2018-205360 5